World of Warcraft: Cataclysm Preview and Interview

After paying a visit to Blizzard Entertainment's campus, the guys at Guardian.co.uk have cranked out both a two-part preview (here and here) and a two-part interview (here and here) for World of Warcraft: Cataclysm. Since the Q&A with Blizzard's Tom Chilton, Greg Street, and Cory Stockton covers more interesting ground, I'll grab a few snips from it:
How much of the current design work is based on watching emergent player activities in the game world and thinking, "hey we should actually support that more fully?"

CS: It totally happens, both from playing the game ourselves and watching other people play. A good example would be Wintergrasp, which was our open-world PvP zone in Northwind. It was the first time we'd ever done something like that, it was PvP but it wasn't in an instance, so any one could come. And it was really crazy, getting it into the game, and it turned out to be one of the most popular parts of the Wrath of Lich King overall.

But we made a massive number of changes to it after it went into the game because people were playing it in very different ways than we expected. We had a system where you had to get a certain amount of honourable kills to get vehicles, but the players ended up doing something completely different to get vehicles so we modified the whole system. The way that they were attacking the bases was way different to how we had planned. The problem with something like that is, with the beta it's hard to get critical mass of people to play it, but when it goes on the server and you have a thousand people going on there at one time, a group mentality works very differently to a small number of players. Definitely, with things like that, we just make updates with every patch. Now we haven't touched it for a while, because it's running exactly how we want.

TC: I'd say most of what we do now is driven by player feedback. It might be feedback we've been hearing since six years ago and we only now have an opportunity to do it, and sometimes it's things that have come up recently. And as you'd expect there are way more ideas than we ever have time to do.

...

So do you find yourself rediscovering game elements that you'd forgotten about?

GS: Oh, all the time! We'll be messing with a class spell balance or something like that and then we'll go, "ah wait, people are using that spell, they've rediscovered it! It's really important!" They'll also find old items that we've forgotten about. It'll be like, "hey this item has a chance on hit to cause a stun". And the item stats alone aren't worth it, but the stun is so powerful they'll continue to use this level 40 item when they're level 80.

There was the notorious sword that caused magic damage rather than physical damage, which we didn't think would be a big problem, but then when you think more about it, you realise how powerful that might be because it bypasses armour, and breaks all of our rules. That was enormous. It's hard, especially as I'm the person who's expected to know more about the class part of the game than, not only any one player, but also the entire community combined. It's impossible. It's easy to forget stuff. And then they're merciless when you publicly display your lack of knowledge.

...

Will we ever have an MMOG that presents players with a single open world and which allows all participants to live and play in that same environment together?

TC: Partly, that's a technological question what happens when we have a server that can support 10,000 people or 50,000 or 100,000 people simultaneously? And then there are the gameplay implications. The game world that we create for WoW can only support a certain number of people in the same place before the gameplay experience becomes degraded. If you're all competing against the same quest monsters, it isn't a lot of fun when there are 1,000 players standing around waiting to whack every single one. So there's only so much that the actual content can support.

So if you were to build a game to be the way you're describing, you'd have to design the content in a way that would expect that and that can have a lot of weird implications for how you make the game. It's also very difficult to make predications about how many people are going to play a game. So if you set out to create a world that could support a million players, how does that world feel when only 100,000 people show up to play? The experience they have might not be as good as you were expecting because you thought there were going to be a million.